The Statistical Center (SC) of the Southeastern Cancer Study Group (SECSG) will provide the statistical analysis and design of SECSG protocol studies. Before a study is activated, the SC will a) specify the number of evaluable cases necessary to achieve statistical significance, and b) estimate the length of time necessary to complete the study. When a patient enters on study, the SC assigns him to a particular treatment group based on randomization or stratification criteria in the protocol. When necessary the SC will implement minimization techniques for patient assignment. The SC will collect relevant initial data for all patients entered on SECSG studies. When a patient completes a study, the SC evaluates his course of treatment and decides, in consultation with Project Chairmen, the evaluability and response to treatment of each patient in accordance with SECSG criteria. All data relevant to a case is to be stored and may be retrieved for analysis. All interim and final analyses of protocols are presented at Group meetings. Such analyses will include rate and duration of response or DFI, survival or response. At such presentations, the SECSG Statisticians discuss the feasibility of continuing or closing a particular study; also, progress reports are prepared for each active protocol. When a study is complete, the statistician assigned to the protocol shall assist the Project Chairmen in preparing a report of the results of the study for publication. The SC will monitor all protocols for unanticipated toxicity or response and, when necessary, report its findings to the Project Chairmen. Every month, the SC compiles a member-institution report listing the number of cases entered, completed and evaluable, and notifies the SECSG Executive Committee of these findings when the committee meets.